Abstract: 

Product categories are the structural backbone of every online shop. To attract customers and facilitate navigation, categories need to be easily understandable, logical and consistent. With the explosive growth of data, it is becoming more and more difficult for retailers to match products to appropriate categories, and large product catalogs as well as the need to quickly adapt to changes often lead to costly misclassifications. In this talk, I will present our approach at commercetools to build a category recommendation system using methods from machine learning. I will talk about our use of deep neural nets and transfer learning to build an image classifier, word2vec and tf-idf to build a text classifier, and how we integrated these models in a REST API.

Bio: 

Amadeus Magrabi is a data scientist at commercetools. He develops machine learning applications and data science solutions to innovate automated product information management and personalized customer experiences. His scientific background is in the area of machine learning and neuroscience.